Macromolecular recognition is one of the fundamental processes that drive complex behavior in biological systems. Understanding the basis for specificity of macromolecular interactions continues to be an important and unsolved problem in biomedical research. Macromolecular interactions between pathogen ligands and host receptors is a particularly useful model for understanding molecular recognition. An extraordinary host-pathogen system for addressing the issue of specificity was recently discovered in a lysogenic bacteriophage of Bordetella. This phage can change its host tropism by varying the sequence of a host receptor-binding protein, called the major tropism determinant (Mtd, 40 kD). These sequence variations involve a small number (~10) of substitutions that are introduced by a genetic mechanism at discrete positions in a 44-residue variable region (VR) of Mtd. It has been calculated that greater than 10^12 unique variants can be created by this process. Selection studies have shown that some of these variants have altered receptor specificities and result in altered host tropism for the phage. At present, no structural data is available to explain how sequence alterations in the variable region of Mtd are tolerated and lead to the formation of a wide array of receptor binding specificities. To address this problem and to test the hypothesis that sequence variations in Mtd change chemical characteristics but not structure, I propose to carry out X-ray crystallographic studies of four Mtd variants that have different receptor specificities. Diffracting crystals of these variants have been obtained, demonstrating the feasibility of this study.